International Journal of
Environmental Research
and Public Health
Article
Openness to Experience Moderates the Association of Warmth
Profiles and Subjective Well-Being in Left-Behind and
Non-Left-Behind Youth
Yongfeng Ma 1 , Chunhua Ma 1 and Xiaoyu Lan 2, *






Citation: Ma, Y.; Ma, C.; Lan, X.
Openness to Experience Moderates
the Association of Warmth Profiles
and Subjective Well-Being in
Left-Behind and Non-Left-Behind
Youth. Int. J. Environ. Res. Public
Health 2022, 19, 4103. https://
doi.org/10.3390/ijerph19074103
Academic Editors: Fernando García
and Paul B. Tchounwou
Received: 4 January 2022
Accepted: 29 March 2022
Published: 30 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Int. J. Environ. Res. Public Health 2022, 19, 4103. https://doi.org/10.3390/ijerph19074103
1 College of Educational Science and Technology, Northwest Minzu University, Lanzhou 730030, China;
mayongfeng@xbmu.edu.cn (Y.M.); mch@xbmu.edu.cn (C.M.)
2 Promenta Research Center, Department of Psychology, University of Oslo, 0373 Oslo, Norway
* Correspondence: xiaoyu.lan@psykologi.uio.no
Abstract: Crouched in the socioecological framework, the present research compared the subjective
well-being of left-behind youth with their non-left-behind peers. Furthermore, this research investi-
gated the association of parental warmth and teacher warmth using a person-centered approach with
adolescents’ subjective well-being on the whole sample, and examined its conditional processes by
ascertaining the moderating role of openness to experience and left-behind status in this association.
A total of 246 left-behind youth (53.6% girls; Mage = 15.77; SD = 1.50) and 492 socio-demographically
matched, non-left-behind peers (55.1% girls; Mage = 15.91; SD = 1.43) was involved in this study.
During school hours, these adolescents were uniformly instructed to complete a set of self-report ques-
tionnaires. The results from ANCOVA exhibited no significant differences in subjective well-being
between these two groups of youth. Moreover, four warmth profiles were revealed: congruent low,
congruent highest, congruent lowest, and incongruent moderate, and youth within the congruent
highest profile were more likely than the other three profiles to report higher subjective well-being.
Additionally, moderation analyses demonstrated that high openness was one protective factor for
subjective well-being, when left-behind youth perceived the lowest levels of parental warmth and
teacher warmth congruently. These findings indicate that left-behind youth may not be psychologi-
cally disadvantaged in terms of positive psychosocial outcomes, such as subjective well-being, and
school activities or social initiatives emphasizing openness to experience would be essential for them
to facilitate positive adaptive patterns after parental migration.
Keywords: subjective well-being; parental warmth; teacher warmth; openness to experience;
left-behind youth
1. Introduction
With the accelerating industrialization process and dramatic economic growth globally,
an unprecedented wave of labor migration has occurred in a few low- or middle-income
yet fast-developing countries from South and East Asia, Africa, and Eastern Europe [1–3].
Individuals who lived in impoverished regions internationally or internally migrated to
developed areas, chasing financially promising job opportunities and better quality of
life [4–6]. Nevertheless, due to financial considerations, many migrated parents have to
leave their children in original communities [7,8]. A plethora of empirical studies have
consistently shown that these left-behind youth who are taken care of by their extended
family members exhibit high emotional and behavioral difficulties due, mainly, to dimin-
ished parental involvement after migration [1,9]. Therefore, examining the correlates of
left-behind youth’s mental health would be very important for developing much-needed,
evidence-based intervention or prevention programs.
In this study, we leveraged Chinese youth as a reference point in the hope of provid-
ing essential insights into worldwide migrant populations and their families. Instead of
https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2022, 19, 4103 2 of 16

focusing on a dominant deficit approach, the current study focused on youths’ strengths
and subjective well-being (SWB), in line with the positive psychology movement [10].
Specifically, SWB refers to an individual’s evaluation of his or her life quality, encom-
passing three key aspects: life satisfaction (cognitive aspect), and positive and negative
affect (emotional aspect) [11,12]. Applying a socio-ecological framework in the present
research [13], we aimed to investigate how youth’s immediate surroundings (e.g., parents
and teachers) and their personal characteristics (e.g., openness to experience) were individ-
ually and jointly associated with SWB. In what follows in the introduction, we review each
of the study variables and present research questions and tested hypotheses, starting from
the presentation of parental warmth and teacher warmth.

1.1. Parental Warmth and Teacher Warmth Using a Person-Centered Approach

In this study, we focused particularly on parents and teachers, as they represent two
prominent sources of social support in youth’s immediate surroundings and are responsive
to youth’s needs, which could be helpful for developing more practical and effective
strategies in promoting youth’s SWB [14,15]. Specifically, parental warmth and teacher
warmth are defined as the quality of the affectional bond that reflects verbal and non-verbal
interactions (e.g., expressions of affection and emotional support) between parents/teachers
and youth, in line with acceptance-rejection theory [16,17]. Several empirical studies
have pointed out that affectionate and warm parents/teachers–youth interaction plays a
prominent role in a youth’s healthy emotional development and better SWB [18,19]. Indeed,
immediate surroundings with high responsiveness and warmth significantly influence
positive emotional climate with high emotional security and a sense of comfort, contributing
to adolescents’ positive social and emotional development [20–22].
Despite these shreds of solid evidence, several conceptual and methodological limita-
tions still merit further investigation. First, there is a shortage of studies that simultaneously
incorporate father warmth, mother warmth, and teacher warmth and investigate their
congruent impacts on adolescents’ SWB [23]. This limitation is particularly salient in the
context of Chinese culture, as, within one family, the roles of father and mother in disciplin-
ing their children can be complementary or significantly distinctive, such as the patterns of
“strict father and kind mother” [24,25]. Additionally, the salient role of teachers in students’
socioemotional functions cannot be ignored, as teachers in each classroom have broad
responsibilities for adolescents’ academic and daily life activities [26,27]. To gain a more
comprehensive understanding of contextual warmth on adolescents’ SWB, it is valuable
and informative to incorporate all these three figures simultaneously in one investigation.
Second, most quantitative studies to date have heavily employed a traditional variable-
centered approach to study the association of isolated warmth figures with adolescents’
SWB. Built upon the assumption that the population is homogeneous [28], the variable-
centered approach fails to consider that adolescents may simultaneously perceive a distinct
degree of warmth from different social agencies, such as parents and teachers, and makes
it challenging to have a holistic perspective of contextual warmth based on natural con-
figurations in adolescents’ social spheres [29,30]. Due to these inherently methodological
limitations, some scholars argue that research should be augmented with a person-centered
approach, assuming that there exist numerous unobserved subgroups within a certain
population [24,28,31]. Adopting a person-centered approach would allow researchers to
explore naturally occurring sub-groups of adolescents combining the differential degree of
perceived father warmth, mother warmth, and teacher warmth, and to discern the associa-
tions of these emerging warmth profiles with SWB for specific subgroups of adolescents.
In concert with the socio-ecological framework [13], we also aimed to demonstrate
whether personality traits, such as openness to experience, may explain conditional pro-
cesses of contextual warmth with adolescents’ SWB, as we discussed below.
Int. J. Environ. Res. Public Health 2022, 19, 4103 3 of 16

1.2. Openness to Experience

Openness to experience refers to individual differences in the proneness to novelty,
tolerance of ambiguity, and openness to new ideas [32,33]. Specifically, adolescents with
greater openness tend to appreciate novel ideas, actively seek unfamiliar experiences, and
experience positive and negative emotions more intensely [32,34]. Although previous stud-
ies have highlighted the critical correlates of personality traits, mainly through extraversion
and neuroticism, with resilience and SWB, these studies also indicate that openness is an
additional important, yet less studied, variable related to SWB [32,35,36].
Importantly, accumulating research has emphasized the vital role of openness in
stress regulation, migrants’ intercultural effectiveness, and their successful sociocultural
adaptation [37–40]. Adolescents with greater openness are willing to try different activi-
ties, process novel experiences, and life changes, representing a valuable prerequisite for
negotiating life changes after parental migration and demonstrating a successful socio-
cultural adaptation [37,38,41]. Given that adolescents who score high on openness are
more appreciative of and responsive to family changes and accept new experiences [42,43],
high open adolescents are likely to respond more positively to parental migration and/or
low contextual warmth. Therefore, it is essential to investigate whether openness to ex-
perience moderates the association between warmth profiles and SWB in left-behind and
non-left-behind youth.

1.3. The Current Study

The present study investigated the following research questions (RQ). RQ1: do left-
behind and non-left-behind youth differ in SWB? As parental migration and subsequent
changes in adolescent living arrangements generate significant stress into their life [44,45],
we anticipated that left-behind youth might report a lower level of SWB than their non-left-
behind peers.
RQ2: Is it possible to identify distinct warmth profiles based on three salient figures
(i.e., fathers, mothers, and teachers) using a mixed sample of left-behind and non-left-
behind youth? Due to the scarcity of literature concerning warmth profiles based on these
three figures, we did not generate a priori hypothesis about the number of warmth profiles
that may emerge. However, according to the previous study [23], we expected that the
presence and the absence of father/mother warmth could co-occur in combination with the
presence and the absence of teacher warmth, thereby producing distinct combinations of
congruent/incongruent warmth profiles across these three figures.
RQ3: Are these emerging warmth profiles linked to SWB? Additionally, are these
expected associations moderated by openness to experience and/or left-behind status?
Given the notable roles of both parents and teachers in adolescents’ SWB [14,15], we
hypothesized that a congruent high warmth profile (compared with other congruent low
or incongruent patterns) would be more strongly and positively associated with SWB.
Moreover, due to parental migration and subsequently diminished capacity to provide
direct and sufficient warmth for left-behind youth, high open adolescents may react more
positively to decreasing contextual warmth and/or left-behind experiences, and thus
report high levels of SWB [41]. Therefore, we expected that openness to experience might
buffer against the detrimental effect of low contextual warmth on SWB, which might be
particularly pronounced for left-behind youth than their non-left-behind peers.
By addressing these research questions, the current study significantly contributes to
extant research in several innovative ways. First and foremost, in contrast to dominant
literature focusing on adverse outcomes, the current study focuses on youths’ strengths
and SWB, moving away from pathological labeling and fueling the discussion on a more
comprehensive understanding of left-behind youth’s adaption. Second, to complement
a traditional variable-centered approach, the current study is among the first using a
person-centered approach to simultaneously incorporate three significant figures in youth’s
social spheres, documenting the association of distinct warmth profiles with youth’s SWB.
Third, the present research contributes to the scientific understanding of the association
Int. J. Environ. Res. Public Health 2022, 19, 4103 4 of 16

under investigation by innovatively documenting the moderating role of openness to

experience and left-behind status therein. This study, therefore, helps to deepen the
theoretical understanding of the interrelated associations between youth’s social spheres
and their personal characteristics, and provides tangible evidence on targeted intervention
or prevention programs aimed at facilitating positive adaptive patterns for youth with
left-behind experiences.

2. Methods
2.1. Participants and Procedures
The current study was based on a cross-sectional design, which was well-suited to
address specific research questions under investigation. Prior to data collection, study
protocols and consent forms were ethically reviewed by the responsible research ethics
committee, ensuring that this study complies with strict ethical standards included in the
Declaration of Helsinki for human rights. School principals and head teachers collaborating
with our research team approved this project, and informed consent forms were received
from parents and adolescents. During class hours, trained research assistants together with
head teachers administrated this investigation, and adolescents were asked to collectively
complete a set of self-report questionnaires in simplified Chinese lasting approximately one
hour, separated by two sessions. During all research processes, adolescents’ participation
in this study was voluntary, and collected data were used solely for scientific purposes.
A total of 2671 Chinese adolescents participated in this investigation, with grade levels
ranging from 7th to 11th (see Supplementary Materials Table S1 for more details). Eligibility
criteria were as follows: (a) adolescents attended public schools and aged between 13 to
18 years old, and (b) for left-behind youth, one or both of their parents migrated to other
cities in China for working continually for at least six months. After applying these criteria,
246 left-behind youth (53.6% girls; Mage = 15.77; SD = 1.50) were identified. Among them,
23.1% of left-behind youth reported that both of their parents migrated to other urban
regions for work, and 68.3% and 8.6% of left-behind youth reported that one of their fathers
and mothers migrated. The duration of their parental migration ranged from 1 to 10 years
(M = 5.6 years). Most of their fathers (63.2%) and mothers (61.7%) achieved a secondary
school education background.
When comparing two independent groups, unequal sample sizes often affect the
robustness of the equal variance assumption in equivalence tests, such as ANOVA, leading
to invalid statistical inferences [46]. This is particularly pronounced, as samples become
increasingly unequal. To bear this in mind, the current study, therefore, focused on the
trade-offs between balanced sample size and sufficient statistical power. Although having
precisely equal numbers of participants for each group would be an ideal solution, it is
possibly difficult to detect potential effects considering the relatively small sample size in
the targeted group (n = 246). In this perspective, we intentionally doubled the number
of non-left-behind youth, following a few empirical studies comparing left-behind with
non-left-behind youth [47–49]. Notably, it is feasible to have unequal sample sizes in
experimental and developmental studies [50]. It has been suggested that the potential bias
caused by this relatively unequal sample size (e.g., up to two times in the control group)
can be neglectable if the whole sample size is sufficient [46].
Specifically, based on the sample size of left-behind youth, we randomly extracted
492 (55.1% girls; Mage = 15.91; SD = 1.43) non-left-behind youth from the original non-
left-behind data pool. We considered this randomly selected sample representative of the
original non-left-behind sample, as these two samples did not differ significantly in salient
sociodemographic features (i.e., age, gender, and family socioeconomic status) nor in study
variables (see Supplementary Materials Table S2). In addition, we conducted a re-sampling
method; that is, we randomly re-derived an approximately 25% non-left-behind sample
from the original non-left-behind data pool, and the main results presented remained
the same.
Int. J. Environ. Res. Public Health 2022, 19, 4103 5 of 16

Furthermore, we conducted a power analysis to determine whether the sample ana-

lyzed could demonstrate sufficient statistical power concerning the analyses planned [51].
It was estimated that a minimum of 128 participants for ANCOVA was needed to detect
anticipated medium effect size (Cohen’s d = 0.50), whereas a minimum of 218 participants
for multiple regression with covariates included was required to detect anticipated medium
effect size (f 2 = 0.15). Based on our selected sample of left-behind and non-left-behind
youth (N = 738), we had above 95% statistical power (α = 0.05, 1 − β = 0.95) to make
significant statistical inference. Therefore, we concluded that the sample analyzed would
be sufficient to demonstrate our RQs.

2.2. Measures
2.2.1. Subjective Well-Being
For assessing positive and negative affect, we used the Affect Balance Scale [52,53].
This scale comprises 14 items, with 8 per positive affect and 6 per negative affect. Sample
items are “things were going your way (positive affect),” and “very lonely or remote from
other people (negative affect).” Participants reported each item on a 4-point scale from
1 (never) to 4 (always). The mean score for each facet was calculated, with a higher score
representing a higher level of a particular affect.
For assessing life satisfaction, we used the Multidimensional Students’ Life Satisfaction
Scale [54,55]. This scale consists of 25 items measuring adolescents’ satisfaction in five
important domains: family (5 items), friend (5 items), school (5 items), living environment
(5 items), and self (5 items). For example, one item is, “my friends treat me well (on the
friend domain).” Adolescents indicated his or her degree of agreement with each item on
a 4-point scale from 1 (totally disagree) to 4 (totally agree). In the present study, these five
subscales were significantly interrelated in both left-behind (r ranged from 0.36 to 0.56) and
non-left-behind youth (r ranges from 0.35 to 0.60). Concerning these moderate associations
among subscales, an average score across these five domains represented adolescents’
multidimensional life satisfaction [56,57].
The current study focused on an aggregated definition that constitutes SWB as adoles-
cents’ overall evaluation of an affect balance and life satisfaction in different dimensions.
Therefore, following the procedure of previous studies [58,59], the total score of SWB was
calculated in two steps. First, we standardized the scores of positive affect, negative affect,
and life satisfaction in the whole sample. Second, we subtracted negative affect from the
sum score of positive affect and life satisfaction, with a high score representing a greater
degree of SWB in the current study.

2.2.2. Parental Warmth

For assessing parental warmth, we used the subscales of the Parental Acceptance-
Rejection/Control Questionnaire [22]. A total of 16 items were used to measure parental
warmth, with 8 per father warmth and 8 per mother warmth. One item example is, “My
mother/father lets me know she/he loves me”. A 5-point Likert scale from 1 (almost never
true) and 5 (almost always true) was applied to assess each item. The mean of these items
separated by each dimension was calculated as the score of father warmth and mother
warmth. A higher score represented higher father warmth and mother warmth.

2.2.3. Teacher Warmth

For assessing teacher warmth, we used one of the subscales of the Children’s Ap-
praisals of Teacher Support [60]. The warmth subscale comprises seven items that measure
adolescents’ perception of teacher support, encouragement, and acceptance. One item
example is, “My teacher likes me”. A 5-point Likert scale from 1 (completely disagree) to
5 (completely agree) was used. We calculated the mean score of these items, with higher
scores representing higher teacher warmth.
Int. J. Environ. Res. Public Health 2022, 19, 4103 6 of 16

2.2.4. Openness to Experience

For assessing openness to experience, we used one of the subscales of the Big Five
Inventory [61,62]. This subscale encompasses ten items, and one example item is, “I see
myself as someone who is original, comes up with new ideas”. The items are scored
on a 5-point Likert scale with anchors from 1 (strongly disagree) to 5 (strongly agree). The
mean score of the 10 items was calculated, with a higher score indicating greater openness
to experience.

2.2.5. Left-Behind Status

Youth were asked, “did your father and/or mother migrate to other cities working
continually for a long period (at least six months)”? Those youth who reported parental
migration were asked additional items to indicate the duration of parental migration and
current residential status. These questions were adapted from Zhao et al. [63].

2.2.6. Confounding Variables

Participants were asked to provide information on their age, gender, parents’ edu-
cation, parents’ occupations, and family monthly income. Following prior research [64],
family SES in this study was represented by a standardized, composite score of parents’
education, parents’ occupations, and family monthly income. These socio-demographic
variables were controlled for because prior research has documented the potential as-
sociation of these variables with SWB [65]. Moreover, adolescents answered a 16-item
questionnaire to assess their social desirability due to the concern about response artifact of
social desirability to self-report questionnaires [66].

2.3. Data Analyses

Data analyses were carried out in SPSS 21.0 [67], Mplus 7.0 [68], and Jamovi 2.3 [69].
The missing values in the current dataset were completely at random [70]. They, therefore,
were replaced based on the mean score on each of the measurements. Prior to testing
research questions and hypotheses, we calculated means and standard deviations for
variables of interest to define the characteristics of the present research, and analyzed
bivariate correlations between variables.
In terms of RQ1, we used an Analysis of Covariance (ANCOVA) to compare the SWB
in two groups of adolescents. In terms of RQ2, we applied a latent profile analysis to
explore warmth profiles. The ideal profile solution was selected based on the comparison
of model fit indices across distinctive profile solutions, such as low AIC, BIC, aBIC, a sig-
nificant likelihood ratio test, as well as high entropy [28]. In terms of RQ3, we applied a
linear regression analysis, statistically controlled for age, gender, family SES, and social
desirability, to examine study associations. In the case of significant two- or three-way
interactions in the liner regression, we leveraged mixed Analysis of Variance (ANOVA),
coupled with plotted figures, to interpret these interactive patterns, because there were
two categorical variables (i.e., warmth profiles and left-behind status) embedded inside
the interaction terms. In this context, left-behind status was regarded as a between-subject
factor, and warmth profiles and openness to experience were treated as within-subject
factors [71]. To increase the accuracy in parameter estimation, we also presented confidence
intervals in terms of unstandardized regression coefficients in the linear regression [72].

3. Results
3.1. Descriptive Statistics and Correlational Analyses
Descriptive statistics and bivariate correlations among study variables, as well as
the reliability coefficients (i.e., Cronbach’s alpha) per each measurement, are presented in
Tables 1 and 2, separated for left-behind and non-left-behind youth.
Int. J. Environ. Res. Public Health 2022, 19, 4103 7 of 16

Table 1. Descriptive statistics and intercorrelations among variables for left-behind youth.

M SD Range Alpha 1 2 3 4 5 6 7 8 9 10 11 12
1. Father warmth 3.58 0.95 1–5 0.89 -
2. Mother warmth 3.83 0.88 1–5 0.87 0.69 *** -
3. Teacher warmth 3.73 0.81 1–5 0.88 0.16 * 0.13 * -
4. Openness 3.78 0.71 1–5 0.80 0.15 * 0.13 * 0.24 *** -
5. Positive affect 2.85 0.35 1–4 0.81 0.10 0.05 0.15 * 0.30 *** -
6. Negative affect 2.18 0.38 1–4 0.78 0.10 −0.04 −0.13 * −0.19 ** −0.46 *** -
7. Life satisfaction 3.16 0.47 1–4 0.92 0.56 *** 0.50 *** 0.28 *** 0.39 *** 0.24 *** −0.09 -
8. Age 15.77 1.50 13–18 - 0.10 0.13 * 0.05 0.11 0.11 −0.13 * 0.09 -
9. Gender a - - 1–2 - 0.08 0.01 0.02 0.02 0.10 −0.17 ** 0.11 0.07 -
10. Socioeconomic
16.18 1.96 8–23 - 0.11 0.07 −0.12 0.02 0.01 −0.06 0.05 0.05 0.09 -
status
11. Years 6.99 3.90 1–17 - −0.04 0.01 −0.03 −0.01 0.09 −0.06 −0.03 0.23 *** 0.12 0.06 -
12. Social
5.25 0.82 1–7 0.85 0.45 *** 0.33 *** 0.29 *** 0.50 *** 0.31 *** −0.17 ** 0.70 *** 0.09 0.12 0.09 −0.02 -
desirability
Note. N = 246. a coded as 1 = boys and 2 = girls. * p < 0.05, ** p < 0.01, *** p < 0.001.

Table 2. Descriptive statistics and intercorrelations among variables for non-left-behind youth.

M SD Range Alpha 1 2 3 4 5 6 7 8 9 10 11
1. Father warmth 3.75 0.87 1–5 0.88 -
2. Mother warmth 3.91 0.80 1–5 0.87 0.76 *** -
3. Teacher warmth 3.83 0.83 1–5 0.90 0.28 *** 0.32 *** -
4. Openness 3.72 0.66 1–5 0.77 0.29 *** 0.29 *** 0.17 *** -
5. Positive affect 2.85 0.36 1–4 0.84 0.22 *** 0.26 *** 0.11 ** 0.11 ** -
6. Negative affect 2.17 0.42 1–4 0.77 −0.04 −0.09 * −0.07 −0.02 −0.49 *** -
7. Life satisfaction 3.26 0.45 1–4 0.93 0.58 *** 0.57 *** 0.40 *** 0.34 *** 0.28 *** −0.07 -
8. Age 15.91 1.43 13–18 - 0.01 −0.01 −0.01 −0.07 −0.03 −0.07 −0.08 -
9. Gender a - - 1–2 - 0.06 0.05 0.04 −0.13 ** 0.17 *** −0.13 ** 0.06 −0.01 -
10. Socioeconomic
15.95 1.95 9–25 - 0.10 * 0.09 * 0.01 0.04 −0.04 0.03 0.05 −0.09 * −0.08 -
status
11. Social
5.27 0.82 1–7 0.86 0.44 *** 0.43 *** 0.35 *** 0.41 *** 0.33 *** −0.16 *** 0.62 *** −0.04 0.06 0.05 -
desirability
Note. N = 492. a coded as 1 = boys and 2 = girls. * p < 0.05, ** p < 0.01, *** p < 0.001.
Int. J. Environ. Res. Public Health 2022, 19, 4103 8 of 16

3.2. RQ1: Group Differences in Subjective Well-Being

The mean score of SWB for left-behind youth was −0.14 (SD = 2.11) and was 0.07
Int. J. Environ. Res. Public Health 2022,(SD
19, x= 2.18) for their non-left-behind peers (see Figure 1). The results based on ANCOVA,
8 of 16
after controlling for age, gender, family SES, and social desirability, exhibited that SWB
were not significantly different in these two groups of adolescents, F (1, 736) = 1.33, p = 0.24.

Figure 1. Group differences in subjective well-being. Note. N = 738. Points—Raw data; Bar/Line—Mean;
Figure 1. Group
Bean—Data differences
distribution; interval. Note. N = 738. Points—Raw data; Bar/Line—
in subjective well-being.
Band—Confidence
Mean; Bean—Data distribution; Band—Confidence interval.
3.3. RQ2: Identifying Warmth Profiles
3.3. RQ2:
Model Identifying
fit indicesWarmth Profileswith one to five latent warmth profiles are presented in
for solutions
TableModel
3. fit indices for solutions with one to five latent warmth profiles are presented
in Table 3.
Table 3. Model fit indices for different warmth profiles.

Table 3. Model fit indices for different warmth profiles. Smallest

AIC BIC aBIC Entropy LMR-LRT BLRT
Profiles (%)
Smallest
AIC BIC aBIC Entropy LMR-LRT BLRT
1-Profile 5593.34 5620.96 5601.91 - - - Profiles- (%)
2-Profile 5068.75 5114.79 5083.04 0.80 513.16 *** 532.58 *** 34.2%
1-Profile 5593.34 5620.96 5601.91 - - - -
3-Profile 4909.64 4974.09 4929.64 0.81 161.02 *** 167.11 *** 8.8%
2-Profile
4-Profile 5068.75
4840.28 5114.79
4923.15 5083.04
4865.99 0.80
0.79 513.16
74.54***
* 532.58
77.36 ***
* 34.2%
7.1%
3-Profile
5-Profile 4909.64
4732.41 4974.09
4833.70 4929.64
4763.84 0.81
0.75 161.02
115.64***
* 167.11
120.01***
* 8.8%
2.6%
4-Profile 4840.28 4923.15 4865.99 0.79 74.54 * 77.36 * 7.1%
Note. N = 738. AIC, BIC, and aBIC are model information criteria, and a lower score indicates a better model
5-Profile 4732.41 4833.70
fit; entropy refers to the4763.84 0.75with a higher
classification accuracy, 115.64
score*representing
120.01
better* classification
2.6%
accuracy;
LMR-LRT and BLRT are two likelihood ratio tests, and a significant p-value indicates that a model with k + 1
Note. Nfits
profiles = 738. AIC, BIC,
significantly andthan
better aBIC are model
a model with kinformation criteria,
profiles. Boldface and
values a lower
refer score indicates
to the optimal a this
solution in
better* model
study. p < 0.05,fit;
***entropy
p < 0.001.refers to the classification accuracy, with a higher score representing bet-
ter classification accuracy; LMR-LRT and BLRT are two likelihood ratio tests, and a significant p-
valueAsindicates that a model
summarized with
in Table + 1 likelihood
3,kthe profiles fitsratio
significantly better than
tests remained a model with
significant k profiles.
for all profile
Boldface values refer to the optimal solution in this study. * p < 0.05, *** p < 0.001.
solutions. We first excluded the five-profile solution, as the smallest percentage of this
profile only represented less than 5% of the whole sample [73]. Of the remaining profile
As summarized
solutions, in Table
the four-profile 3, the likelihood
achieved lower AIC,ratioBIC,tests remained
adjusted BIC, significant
and almostfor all pro-
identical
file solutions.
degree We first excluded
of classification accuracy,theas five-profile
demonstrated solution, as the values.
by entropy smallestTherefore,
percentageoverall
of this
profile only
speaking, represented
these model fitless thanfavored
indices 5% of the thewhole samplesolution
four-profile [73]. Ofthat
the was
remaining profile
employed to
solutions,
analyze the four-profile achieved lower AIC, BIC, adjusted BIC, and almost identical
subsequently.
degree of classification
Figure 2 depicts theaccuracy,
structure asof demonstrated
these emergingbyfour entropy
warmthvalues. Therefore,
profiles based onoverall
stan-
speaking, these model fit indices favored the four-profile solution that was employed
dardized scores for each indicator. These profiles were labeled according to Chung et to
al.’s
analyze subsequently.
research [23]. Specifically, youth in the first profile (n = 157; 21.3%), named as “congruent
low,”Figure
reported2 depicts the structure
low scores on fatherofwarmth,
these emerging
mother four warmth
warmth, and profiles
teacher based
warmth;on stand-
youth
ardized
in scores
the second for each
profile (n =indicator. These
267; 36.2%), profiles
labeled were labeled
as “congruent according
highest,” to Chung
reported et al.’s
the highest
research [23]. Specifically, youth in the first profile (n = 157; 21.3%), named as “congruent
low,” reported low scores on father warmth, mother warmth, and teacher warmth; youth
in the second profile (n = 267; 36.2%), labeled as “congruent highest,” reported the highest
scores on father warmth, mother warmth, and teacher warmth; youth in the third profile
(n = 54; 7.3%), named as “congruent lowest,” displayed the lowest scores on father
Int. J. Environ. Res. Public Health 2022, 19, 4103 9 of 16

Int. J. Environ. Res. Public Health 2022, 19, x 9 o

scores on father warmth, mother warmth, and teacher warmth; youth in the third pro-
file (n = 54; 7.3%), named as “congruent lowest,” displayed the lowest scores on father
warmth, motherlabeled as “incongruent
warmth, and teachermoderate,” showed
warmth; youth low-to-moderate
in the scores
fourth profile (n on35.2%),
= 260; father and teac
labeled as “incongruent moderate,” showed low-to-moderate scores on father and teacher in fat
warmth, but moderate-to-high scores on mother warmth. Mean differences
warmth, butwarmth, mother warmth,
moderate-to-high scoresand
on teacher
mother warmth
warmth.across
Meanthese four profiles
differences are represen
in father
in the Supplementary Materials Table S3.
warmth, mother warmth, and teacher warmth across these four profiles are represented in
the Supplementary Materials Table S3.

Figure 2. Four warmth profiles based on standardized scores of father warmth, mother warmth, and
Figure 2. Four warmth profiles based on standardized scores of father warmth, mother warm
andNote.
teacher warmth. N =warmth.
teacher 738. Note. N = 738.
3.4. RQ3: Associations of Warmth Profiles, Openness to Experience, Left-Behind Status with
3.4. RQ3: Associations of Warmth Profiles, Openness to Experience, Left-Behind Status with
Subjective Well-Being
Subjective Well-Being
Before conducing multiple linear regression, we checked whether the data met each of
the assumptions Before conducing
of this analysis, andmultiple linear regression,
the corresponding resultswe checked
have whether in
been reported thethe
data met e
of the assumptions
Supplementary Materials. of this analysis, and the corresponding results have been reported
Table 4 the Supplementary
presents Materials.
the results of the linear regression analysis, F (19, 717) = 22.3, p < 0.001,
AIC = 2926, BIC Table
= 3022,4 RMSE
presents the results
= 1.71. In this of the linear
analysis, we regression
treated theanalysis,
congruentF (19, 717) = 22.3,
highest
profile as the0.001, AICgroup,
reference = 2926,and
BIC = 3022, RMSE
compared each of=the
1.71. In three
other this analysis, we treated
warmth profiles the congru
to this
highesttotally
group. The model profileexplained
as the reference
37.2% ofgroup, and compared
the variance of SWB.each of the other three warmth p
files to this group. The model totally explained 37.2% of the variance of SWB.
Table 4. Regression analysis predicting subjective well-being from warmth profiles, openness to
Table
experience, and 4. Regression
left-behind status. analysis predicting subjective well-being from warmth profiles, opennes
experience, and left-behind status.
Variables b b SE 95% CI for b t p
Variables b b SE 95% CI for b t p
Congruent low −1.02 Congruent 0.21 low −1.43 −0.61−1.02 −
0.21
4.89 −1.43 −0.61 −4.89 <0.0
<0.001
Congruent lowest −1.29 Congruent 0.29 lowest −1.86 −0.73−1.29 −4.47 −1.86 <0.001
0.29 −0.73 −4.47 <0.0
Incongruent moderate −0.45 0.17 −0.79 −0.12 −2.66 0.01
Incongruent moderate −0.45 0.17 −0.79 −0.12 −2.66 0.0
Openness 0.37 0.13 0.12 0.62 2.93 0.01
Openness 0.37 0.13 0.12 0.62 2.93 0.0
Left-behind status a 0.02 0.17 −0.31 0.35 0.10 0.92
Age 0.03 Left-behind
0.04 status a
−0.05 0.120.02 0.17
0.78 −0.31 0.35
0.43 0.10 0.9
Gender b 0.53 Age
0.13 0.27 0.790.03 0.04
3.98 −0.05 0.12
<0.001 0.78 0.4
Socioeconomic status −0.04 Gender
0.04 b
−0.11 0.030.53 0.13
−1.24 0.27 0.79
0.22 3.98 <0.0
Social desirability 1.01 Socioeconomic
0.09 status 0.83 1.20−0.04 0.04
1.07 −0.11 <0.001
0.03 −1.24 0.2
Congruent low × openness −0.03 Social desirability
0.28 −0.59 0.521.01 0.09
−0.12 0.83 0.901.20 1.07 <0.0
Congruent lowest × openness Congruent 0.38
0.46 low x openness
−0.29 1.20−0.03 0.28
1.20 −0.59 0.23
0.52 −0.12 0.9
Incongruent moderate × openness −Congruent
0.07 lowest −0.55
0.24 x openness 0.410.46 −0.29 −0.29 0.77
0.38 1.20 1.20 0.2
Congruent low × left-behind status 0.46 0.40
Incongruent moderate x openness− 0.32 1.24−0.07 1.15
0.24 −0.55 0.25
0.41 −0.29 0.7
Congruent low x left-behind status 0.46 0.40 −0.32 1.24 1.15 0.2
Congruent lowest x Left-behind status 1.63 0.55 0.55 2.71 2.95 0.0
Incongruent moderate x left-behind status 0.99 0.33 0.34 1.63 2.97 0.0
 Int. J. Environ. Res. Public Health 2022, 19, 4103 10 of 16

Table 4. Cont.
Int. J. Environ. Res. Public Health 2022, 19, x 10 of 16
Variables b b SE 95% CI for b t p
Congruent lowest × Left-behind status 1.63 0.55 0.55 2.71 2.95 0.01
Incongruent moderate × left-behind status 0.99 0.33 0.34 1.63 2.97 0.01
Openness x left-behind status 0.90 0.24 0.43 1.36 3.78 <0.001
Openness × left-behind status 0.90 0.24 0.43 1.36 3.78 <0.001
Congruent
Congruent low × openness × left-behind statuslow x openness
−0.76 x left-behind
0.57 status −0.76
−1.87 0.57 0.35−1.87 0.35
−1.34 −1.34 0.18 0.18
Congruent
Congruent lowest × openness × lowest x openness x left-behind status 1.88 0.76 0.39 3.38 2.47 0.01
Incongruent moderate x1.88
left-behind status
0.76
openness x left-behind 0.39 3.38 2.47 0.01
−0.39 0.49 −1.35 0.57 −0.80 0.42
Incongruent moderate × openness × status
−0.39 0.49 −1.35 0.57 −0.80 0.42
Note. N = 738. The reference group for warmth profiles was the congruent highest profile. a coded
left-behind status
asNote.
1 = left-behind
N = 738. The youth and
reference 0 = non-left-behind
group for warmth profilesyouth,
b coded as 1 = boys and 2
was the congruent = girls.
highest profile. a coded as 1 = left-behind
youth and 0 = non-left-behind youth, b coded as 1 = boys and 2 = girls.
As shown in Table 4, adolescents within the congruent highest profile were more
likely than the other
As shown three4,profiles
in Table adolescentsto report
withinhigher SWB. Openness
the congruent highest was positively
profile were more related
likely
tothan
SWB.theMoreover,
other three a significant
profiles tothree-way interaction
report higher (warmth profiles
SWB. Openness (congruent
was positively high-to
related
est vs. congruent
SWB. Moreover,lowest) x openness
a significant three-way x left-behind
interaction status)
(warmthwasprofiles
identified.
(congruent highest vs.
Follow-up post hoc comparisons showed that, in terms of left-behind youth within
congruent lowest) x openness x left-behind status) was identified.
the congruent
Follow-up lowest
postprofile, those with higher
hoc comparisons showedopenness reported
that, in terms higher SWByouth
of left-behind than those
within
with lower openness (b = 2.87, SE = 0.73, t = 3.93, pBonferroni = 0.002). By contrast, for left-
the congruent lowest profile, those with higher openness reported higher SWB than those
with lower
behind youthopenness
within the (b = 2.87, SE =highest
congruent 0.73, t =profile,
3.93, pBonferroni
the scores= 0.002).
of SWBBy contrast,
were for left-
not signifi-
behind youth within the congruent highest profile, the scores of SWB were not
cantly different across distinct levels of openness (b = 0.76, SE = 0.40, t = 1.91, pBonferroni = significantly
different
1.00; across3).distinct
see Figure levelsthat
It indicated of openness wasSE
(b = 0.76,
high openness = 0.40, t = factor
a protective 1.91, pBonferroni = 1.00; see
when left-behind
Figure 3). It indicated that high openness was a protective
youth perceived the lowest levels of parental and teacher warmth congruently. factor when left-behind youth
perceived the lowest levels of parental and teacher warmth congruently.

Figure 3. Interaction effect of warmth profiles and openness on subjective well-being in left-behind
Figure 3. Interaction effect of warmth profiles and openness on subjective well-being in left-behind
youth. Note. N = 246. Openness was divided into three levels based on mean: Mean − 1 SD, Mean,
youth. Note. N = 246. Openness was divided into three levels based on mean: Mean−1 SD, Mean,
and
and Mean +SD.
Mean+1 1 SD.

With regard to non-left-behind youth within both the congruent lowest profile (b = 0.07,
With regard to non-left-behind youth within both the congruent lowest profile (b =
SE = 0.27, t = 0.25, pBonferroni = 1.00) and the congruent highest profile (b = 0.23, SE = 0.67,
0.07, SE = 0.27, t = 0.25, pBonferroni = 1.00) and the congruent highest profile (b = 0.23, SE =
t = 0.34, pBonferroni = 1.00; see Figure 4), the scores of SWB were not significantly different
0.67, t = 0.34, pBonferroni = 1.00; see Figure 4), the scores of SWB were not significantly differ-
across distinct levels of openness. Moreover, as shown in Figure 4, youth within the
ent across distinct levels of openness. Moreover, as shown in Figure 4, youth within the
congruent highest profile were more likely to report higher SWB than those within the
congruent highest profile were more likely to report higher SWB than those within the
congruent lowest profile, regardless of the levels of openness (b = 2.15, SE = 0.53, t = 4.05,
congruent lowest profile, regardless of the levels of openness (b = 2.15, SE = 0.53, t = 4.05,
pBonferroni < 0.001 for higher openness; b = 1.99, SE = 0.66, t = 2.99, pBonferroni = 0.05 for
pBonferroni < 0.001 for higher openness; b = 1.99, SE = 0.66, t = 2.99, pBonferroni = 0.05 for lower
lower openness).
openness).
Int. J. Environ. Res. Public Health 2022, 19, x 11 of 16
Int. J. Environ. Res. Public Health 2022, 19, 4103 11 of 16

Figure 4. Interaction effect of warmth profiles and openness on subjective well-being in non-left-
Figure 4. Interaction effect of warmth profiles and openness on subjective well-being in non-left-
behind youth. Note. N = 492. Openness was divided into three levels based on mean: Mean − 1 SD,
behind youth. Note. N = 492. Openness was divided into three levels based on mean: Mean–1 SD,
Mean,and
Mean, andMean+1
Mean +SD.1 SD.

4. Discussion
4. Discussion
Left-behind youth are rapidly growing and increasingly visible worldwide [1–3], but
theirLeft-behind youth are rapidly
positive psychological growing
correlates haveand increasingly
received visible worldwide
little research attention. [1–3], but
To narrow
their positive psychological correlates have received little
these knowledge gaps in the literature, the current research compared the SWB of left- research attention. To narrow
these
behind knowledge gaps in the youth.
and non-left-behind literature, Thisthe currentmoreover,
research, research compared
examined the the SWB of left-of
association
behind
parental and non-left-behind
warmth and teacher youth.
warmth Thisusing
research, moreover, examined
a person-centered approach thewith
association
SWB on of the
parental
combined warmth
sample, andand teacher warmth whether
investigated using a person-centered
this association was approach with SWB
moderated on the
by openness
combined
to experience sample, and and investigated
left-behind whether
status. this association
Unexpectedly, was moderated
the findings found no by openness
significant
todifferences
experienceinand left-behind status. Unexpectedly, the findings
SWB across these two groups of adolescents. Moreover, four warmth profiles found no significant dif-
ferences in SWB across these two groups of adolescents.
were revealed: congruent low, congruent highest, congruent lowest, and incongruent Moreover, four warmth profiles
were revealed:
moderate. congruent
Adolescents low,the
within congruent
congruent highest,
highestcongruent
profile werelowest, and incongruent
more likely than the other
moderate.
three profiles Adolescents
to report within
higher SWB.the congruent
Additionally,highest profile were
moderation more
analyses likely than
exhibited that the
high
other threewas
openness profiles to report
regarded as one higher SWB.factor
protective Additionally,
for SWB when moderation analyses
left-behind youthexhibited
perceived
that
thehigh
lowestopenness
levels of was regarded
parental warmth as oneandprotective
teacher factor
warmth forcongruently.
SWB when left-behind youth
perceived the lowest
In terms of RQ1, levels
we of parental the
compared warmth
SWB and teacher warmth
of left-behind youthcongruently.
with non-left-behind
In terms
peers. Contraryof RQ1,
to our wefirst
compared
hypothesis the and
SWBthe of literature
left-behind youth with
concerning SWB non-left-behind
of left-behind
peers. Contrary
adolescents to our
in rural first [44,45],
China hypothesis and theshowed
the results literature thatconcerning SWB ofofleft-behind
these two groups adolescents
adolescents
did not differ in rural China
in SWB. [44,45],
This finding the results showedthat
may indicate that“hopes
these two andgroups
hurdles” of adolescents
may coexist
did
in not differ inadaption
adolescent SWB. This finding may
following indicate
parental that “hopes
migration in urbanand China
hurdles” may
[74]. Onecoexist in
possible
adolescent
explanation adaption
could be following
related parental migrationconditions
to socioeconomic in urban China in urban [74]. One possible
China. ex-
Unlike rural
planation could be related
areas in contemporary to socioeconomic
China, urban regions conditions in urban China.
present favorable Unlikeopportunities
educational rural areas
in contemporary China, urban regions present favorable educational opportunities
and socioeconomic conditions [75]. In this context, the possible disadvantage and
of parental
socioeconomic conditionsadolescents’
migrant on left-behind [75]. In this SWB context,
may thebepossible
somehow disadvantage
compensated of parental mi-
[64]. Another
grant on left-behind
possible interpretation adolescents’
could be SWB becausemay be somehow
migration may compensated [64]. Another
economically benefit family pos-
mem-
sible
bersinterpretation
left behind, as could be because
remittance migration
transfers can may
ease economically
budget constraints benefitandfamily members
improve their
left behind, as remittance transfers can ease budget constraints and improve their quality
quality of life. In addition, it is noteworthy that these two groups of adolescents were
ofsocio-demographically
life. In addition, it is noteworthy
matched. The that theselevels
similar two groups
of family of SES
adolescents
and relatedwere socio-de-
demographic
characteristics may potentially weaken the group differences
mographically matched. The similar levels of family SES and related demographic char- in SWB.
Concerning
acteristics RQ2, byweaken
may potentially extending prior research
the group differences of warmth
in SWB. profiles [23], the current
findings revealed
Concerning RQ2,four bywarmth
extending profiles: congruent
prior research of low,
warmth congruent
profileshighest,
[23], thecongruent
current
lowest, and incongruent moderate. Of these profiles,
findings revealed four warmth profiles: congruent low, congruent highest, congruent the congruent highest represented
the largest
lowest, percentage inmoderate.
and incongruent the wholeOfsample. This may
these profiles, theindicate
congruent that highest
parentsrepresented
and teachers
are salient figures in adolescents’ social spheres, and provide
the largest percentage in the whole sample. This may indicate that parents and teachers immense emotional support.
are salient figures in adolescents’ social spheres, and provide immense emotional support.on
However, the measurement per parental warmth and teacher warmth was solely based
self-report questionnaires, and it is possible that adolescents may respond to these items
Int. J. Environ. Res. Public Health 2022, 19, 4103 12 of 16

in a socially desirable manner. Thus, this caution should be kept in mind when readers
explain this finding. Moreover, the second largest percentage of adolescents fall within
the incongruent moderate profile, characterized by moderate-to-high mother warmth yet
low-to-moderate father and teacher warmth. One possible explanation for this pattern
is that mothers are still regarded as the primary caregiver and prominent figures to offer
emotional support in contemporary Chinese society. In addition, the remaining two profiles
were characterized by different degrees of low parental and teacher warmth congruently.
Concerning RQ3, we investigated the association of these emerging warmth profiles
with SWB, as well as its conditional processes. In line with our expectations, adolescents
within the congruent highest profiles are more likely than the other three profiles to report
higher levels of SWB. This finding corroborates prior research [18,22,76], suggesting that
parents and teachers are significant figures in adolescent daily life and contribute jointly
and significantly to their well-being. One possible explanation is that parental warmth
and teacher warmth can provide adolescents with high emotional security and a sense of
support. These positive experiences accelerate the formation of developmental assets, such
as adaptive psychological and behavioral patterns, which enhances adolescents’ SWB [77].
Furthermore, the interactive pattern exhibited that when reporting the lowest lev-
els of parental warmth and teacher warmth congruently, left-behind adolescents with
higher openness to experience are more likely to report higher SWB. Adolescents with
high openness may tend to effectively regulate life-stress and use adaptive coping strate-
gies in response to parental migration [34,78]. Moreover, as demonstrated by previous
studies [79,80], openness is positively related to flexibility and creativity, enabling adoles-
cents to accept or initiate adaptive changes in life. In this regard, left-behind youth with
high openness report greater SWB than low-openness adolescents who are more vulnerable
to parental migration [39].
When interpreting these significant findings, a few notable limitations of this study
should be bear in mind. First, the present research was conducted cross-sectionally, and
thus we cannot infer causality concerning the associations under investigation. Second,
although self-reported questionnaires employed in the current study have been validated in
Chinese adolescents and demonstrated as reasonably appropriate to study these constructs,
we cannot entirely exclude the possibility of common method bias that may inflate study
associations. Third, our findings should be considered within the cultural boundaries of
Chinese society, as the current study was limited by relying on a relatively smaller sample
size of Chinese left-behind and non-left-behind youth. Finally, the current study only
focused on global SWB, and the in-depth exploration of the domain-specific life satisfaction
and emotional aspects would be informative [81].
Despite these limitations, the current study provides substantial theoretical and practi-
cal implications. In terms of theory, the current study enriches the socioecological frame-
work among youth with and without left-behind experiences, demonstrating the interactive
patterns between their immediate surroundings and personal characteristics. With regard
to practical implications, the current study suggests that “hopes and hurdles” may coexist in
youth’s psychosocial adaptation after parental migration. Therefore, researchers interested
in left-behind youth are recommended to employ both positive and negative outcomes to
have a comprehensive understanding of their psychosocial adaption. At the same time,
policymakers should implement a few strategies to help reduce “negative stereotypes”
on left-behind youth, as they are not always psychologically disadvantaged. Moreover,
intervention or prevention strategies should pay specific attention to bolstering support
systems and nurturing warm and supportive relationships between youth and their par-
ents/teachers. For instance, practitioners or educators should initiate professional guidance
during regular parent–teacher meetings with parents and teachers. They can supervise
parents and teachers on how to emotionally express themselves and behaviorally provide
warmth in responses to youth’s needs, equipping them with sufficient skills when youth
approach them for help. Likewise, practitioners or educators have to highlight the essential
role of regular contacts and collaborations between parents and teachers for positive youth
Int. J. Environ. Res. Public Health 2022, 19, 4103 13 of 16

development. For school-aged adolescents, their better SWB depends not only on teachers’
supervision and support but also on a congruent format of joint endeavors between parents
and teachers. Additionally, left-behind youth are advised to receive some training tasks at
school to improve their openness to experience [82,83], ensuring their positive adaption
patterns after parental migration. Practitioners or educators should highlight the variety of
these designed tasks, incorporate adventurous spirits inside, and gradually bump up the
challenges as their skills develop.

5. Conclusions
To recap, the findings advance existing scholarship, indicating that left-behind youth
may not be disadvantaged in terms of positive psychological outcomes, such as SWB.
Perhaps more importantly, for left-behind youth who perceive less contextual warmth
from their immediate environments, school activities or social initiatives emphasizing
openness to experience would be essential for them to develop positive adaptive patterns
after parental migration.

Supplementary Materials: The following supporting information can be downloaded at: https://
www.mdpi.com/article/10.3390/ijerph19074103/s1, Table S1: The number of participants distributed
by eight public schools where the data were collected; Table S2: Mean differences in variables of
interest between selected non-left-behind youth and original non-left-behind data pool; Table S3:
Mean differences in study indicators across four warmth profiles.
Author Contributions: Conceptualization, Y.M. and X.L.; Methodology, Y.M. and X.L.; Formal
Analysis, X.L.; Investigation, Y.M. and C.M.; Resources, Y.M. and C.M.; Data Curation, C.M.; Writing—
Original Draft Preparation, Y.M.; Writing—Review and Editing, X.L.; Project Administration, C.M.;
Funding Acquisition, C.M. and Y.M. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was supported by the General Project of Philosophy and Social Science
Planning of Gansu Province in 2021 (ID: 2021YB022) and the Fundamental Research Funds for the
Central Universities (Northwest Minzu University; Project No. 31920210123, 31920210095, YLKC-13).
Institutional Review Board Statement: The study was conducted in accordance with the Declara-
tion of Helsinki, and approved by the Institutional Review Board of Northwest Minzu University
(approval code XBMU20200720).
Informed Consent Statement: Informed consent was obtained from all participants involved in
this study.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The authors would like to thank voluntary participants and research assistants
for their valuable contribution to this study. At the same time, we appreciate the three anonymous
reviewers’ highly constructive comments that led us to improve the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Antia, K.; Boucsein, J.; Deckert, A.; Dambach, P.; Račaitė, J.; Šurkienė, G.; Jaenisch, T.; Horstick, O.; Winkler, V. Effects of
international labour migration on the mental health and well-being of left-behind children: A systematic literature review. Int. J.
Environ. Res. Public Health 2020, 17, 4335. [CrossRef] [PubMed]
2. Wickramage, K.; Siriwardhana, C.; Vidanapathirana, P.; Weerawarna, S.; Jayasekara, B.; Pannala, G.; Adikari, A.; Jayaweera, K.;
Peiris, S.; Siribaddana, S.; et al. Risk of mental health and nutritional problems for left-behind children of international labor
migrants. BMC Psychiatry 2015, 15, 39. [CrossRef] [PubMed]
3. Fellmeth, G.; Rose-Clarke, K.; Zhao, C.; Busert, L.K.; Zheng, Y.; Massazza, A.; Sonmez, H.; Eder, B.; Blewitt, A.; Lertgrai, W.; et al.
Health impacts of parental migration on left-behind children and adolescents: A systematic review and meta-analysis. Lancet
2018, 392, 2567–2582. [CrossRef]
4. Lei, H.; Zhang, Q.; Wang, Z.; Shao, J. A Longitudinal Study of Depressive Symptoms and Delinquency among Chinese Left-behind
Children. Psychiatry Res. 2021, 301, 113955. [CrossRef]
Int. J. Environ. Res. Public Health 2022, 19, 4103 14 of 16

5. Yang, B.; Xiong, C.; Huang, J. Parental emotional neglect and left-behind children’s externalizing problem behaviors: The
mediating role of deviant peer affiliation and the moderating role of beliefs about adversity. Child. Youth Serv. Rev. 2021,
120, 105710. [CrossRef]
6. Li, Q.; Zhang, W.; Zhao, J. The longitudinal associations among grandparent–grandchild cohesion, cultural beliefs about adversity,
and depression in Chinese rural left-behind children. J. Health Psychol. 2021, 26, 140–155. [CrossRef]
7. Lan, X.; Radin, R. Direct and interactive effects of peer attachment and grit on mitigating problem behaviors among urban
left-behind adolescents. J. Child Fam. Stud. 2020, 29, 250–260. [CrossRef]
8. Wang, F.; Lin, L.; Lu, J.; Cai, J.; Xu, J.; Zhou, X. Mental health and substance use in urban left-behind children in China: A growing
problem. Child. Youth Serv. Rev. 2020, 116, 105135. [CrossRef]
9. Wang, L.; Mesman, J. Child Development in the Face of Rural-to-Urban Migration in China: A Meta-Analytic Review. Perspect.
Psychol. Sci. 2015, 10, 813–831. [CrossRef]
10. Seligman, M.; Csikszentmihalyi, M. Positive psychology: An introduction. Am. Psychol. 2000, 55, 5–14. [CrossRef]
11. Diener, E. Subjective well-being: The science of happiness and a proposal for a national index. Am. Psychol. 2000, 55, 34–43.
[CrossRef]
12. Diener, E.; Lucas, R.E.; Oishi, S. Subjective well-being: The science of happiness and life satisfaction. In The Handbook of Positive
Psychology; Snyder, C.R., Lopez, S.J., Eds.; Oxford University Press: Oxford, UK, 2002; pp. 63–73.
13. Bronfenbrenner, U.; Morris, P.A. The bioecological model of human development. In Handbook of Child Psychology: Volume 1.
Theoretical Models of Human Development; Lerner, R.M., Ed.; Wiley: Hoboken, NJ, USA, 2006; pp. 793–828.
14. Olsson, E. The role of relations: Do disadvantaged adolescents benefit more from high-quality social relations? Acta Sociol. 2009,
52, 263–286. [CrossRef]
15. Chirkov, V.I.; Ryan, R.M. Parent and teacher autonomy-support in Russian and US adolescents: Common effects on well-being
and academic motivation. J. Cross-Cult. Psychol. 2001, 32, 618–635. [CrossRef]
16. Rohner, R.P.; Khaleque, A.; Cournoyer, D.E. Introduction to parental acceptance-rejection theory, methods, evidence, and
implications. J. Fam. Theory Rev. 2012, 2, 73–87. [CrossRef]
17. Epkins, C.C.; Harper, S.L. Mothers’ and fathers’ parental warmth, hostility/rejection/neglect, and behavioral control: Specific
and unique relations with parents’ depression versus anxiety symptoms. Parenting 2016, 16, 125–145. [CrossRef]
18. Khaleque, A. Perceived Parental Warmth, and Children’s Psychological Adjustment, and Personality Dispositions: A Meta-
analysis. J. Child Fam. Stud. 2013, 22, 297–306. [CrossRef]
19. Pianta, R.C. Enhancing Relationships between Children and Teachers; American Psychological Association: Washington, DC, USA, 1999.
[CrossRef]
20. Morris, A.S.; Silk, J.S.; Steinberg, L.; Myers, S.S.; Robinson, L.R. The role of the family context in the development of emotion
regulation. Soc. Dev. 2007, 16, 361–388. [CrossRef]
21. Tennant, J.E.; Demaray, M.K.; Malecki, C.K.; Terry, M.N.; Clary, M.; Elzinga, N. Students’ ratings of teacher support and academic
and social–emotional well-being. Sch. Psychol. Q. 2015, 30, 494–512. [CrossRef]
22. Rohner, R.P. Parental Acceptance-Rejection Questionnaire (PARQ): Test manual. In Handbook for the Study of Parental Acceptance
and Rejection, 4th ed.; Rohner, R.P., Khaleque, A., Eds.; Center for the Study of Parental Acceptance and Rejection, University of
Connecticut: Storrs, CT, USA, 2005; pp. 43–106.
23. Chung, G.; Phillips, J.; Jensen, T.M.; Lanier, P. Parental Involvement and Adolescents’ Academic Achievement: Latent Profiles of
Mother and Father Warmth as a Moderating Influence. Fam. Process 2020, 59, 772–788. [CrossRef]
24. Cui, G.; Lan, X. The associations of parental harsh discipline, adolescents’ gender, and grit profiles with aggressive behavior
among Chinese early adolescents. Front. Psychol. 2020, 11, 323. [CrossRef]
25. Shek, D.T.L. Perceived parental control and parent-child relational qualities in Chinese adolescents in Hong Kong. Sex Roles 2005,
53, 635–646. [CrossRef]
26. Lan, X.; Moscardino, U. Direct and interactive effects of perceived teacher-student relationship and grit on student wellbeing
among stay-behind early adolescents in urban China. Learn. Individ. Differ. 2019, 69, 129–137. [CrossRef]
27. Tian, L.; Tian, Q.; Huebner, E.S. School-related social support and adolescents’ school-related subjective well-being: The mediating
role of basic psychological needs satisfaction at school. Soc. Indic. Res. 2016, 128, 105–129. [CrossRef]
28. Tein, J.Y.; Coxe, S.; Cham, H. Statistical power to detect the correct number of classes in latent profile analysis. Struct. Equ. Model.
A Multidiscip. J. 2013, 20, 640–657. [CrossRef] [PubMed]
29. Bergman, L.R.; von Eye, A.; Magnusson, D. Person-oriented research strategies in developmental psychopathology. In Develop-
mental Psychopathology, 2nd ed.; Cicchetti, D., Cohen, D.J., Eds.; Wiley: London, UK, 2006; pp. 850–888.
30. Weaver, S.R.; Kim, S.Y. A person-centered approach to studying the linkages among parent-child differences in cultural orientation,
supportive parenting, and adolescent depressive symptoms in Chinese American families. J. Youth Adolesc. 2008, 37, 36–49.
[CrossRef]
31. Crockett, L.J.; Moilanen, K.L.; Raffaelli, M.; Randall, B.A. Psychological profiles and adolescent adjustment: A person-centered
approach. Dev. Psychopathol. 2006, 18, 195–214. [CrossRef]
32. McCrae, R.R.; Costa, P.T., Jr. Personality trait structure as a human universal. Am. Psychol. 1997, 52, 509–516. [CrossRef]
33. McCrae, R.R. Openness to Experience. In Encyclopedia of Applied Psychology; Spielberger, C., Ed.; Elsevier: Oxford, UK, 2004;
pp. 707–709. [CrossRef]
Int. J. Environ. Res. Public Health 2022, 19, 4103 15 of 16

34. Wu, T.Y.; Hu, C. Abusive supervision and subordinate emotional labor: The moderating role of openness personality. J. Appl. Soc.
Psychol. 2013, 43, 956–970. [CrossRef]
35. Soto, C.J. Is Happiness Good for Your Personality? Concurrent and Prospective Relations of the Big Five with Subjective
Well-Being. J. Personal. 2015, 83, 45–55. [CrossRef]
36. Oshio, A.; Taku, K.; Hirano, M.; Saeed, G. Resilience and Big Five personality traits: A meta-analysis. Personal. Individ. Differ.
2018, 127, 54–60. [CrossRef]
37. Mak, A.S.; Tran, C. Big five personality and cultural relocation factors in Vietnamese Australian students’ intercultural social
self-efficacy. Int. J. Intercult. Relat. 2001, 25, 181–201. [CrossRef]
38. Ward, C.; Leong, C.H.; Low, M. Personality and sojourner adjustment: An exploration of the Big Five and the cultural fit
proposition. J. Cross-Cult. Psychol. 2004, 35, 137–151. [CrossRef]
39. Williams, P.G.; Rau, H.K.; Cribbet, M.R.; Gunn, H.E. Openness to experience and stress regulation. J. Res. Personal. 2009,
43, 777–784. [CrossRef]
40. Xin, Y.; Wu, J.; Yao, Z.; Guan, Q.; Aleman, A.; Luo, Y. The relationship between personality and the response to acute psychological
stress. Sci. Rep. 2017, 7, 16906. [CrossRef]
41. Danckert, B.; Dinesen, P.T.; Klemmensen, R.; Nørgaard, A.S.; Stolle, D.; Sønderskov, K.M. With an Open Mind: Openness to
Experience Moderates the Effect of Interethnic Encounters on Support for Immigration. Eur. Sociol. Rev. 2017, 33, 721–733.
[CrossRef]
42. Schwaba, T.; Luhmann, M.; Denissen, J.J.A.; Chung, J.M.; Bleidorn, W. Openness to experience and culture-openness transactions
across the lifespan. J. Personal. Soc. Psychol. 2018, 115, 118–136. [CrossRef]
43. Whitbourne, S.K. Openness to experience, identity flexibility, and life change in adults. J. Personal. Soc. Psychol. 1986, 50, 163–168.
[CrossRef]
44. Chen, Y.; Wang, L.; Zhao, J. Peer relationship profiles in rural Chinese adolescents: Longitudinal relations with subjective
well-being. J. Health Psychol. 2019, 26, 1803–1820. [CrossRef]
45. Ye, L.; Qian, Y.; Meng, S.; Ye, D.; Rong, C.; Vandenhouten, E.E.; Jing, F.; Mao, Y. Subjective well-being of left-behind children:
A cross-sectional study in a rural area of eastern China. Child Adolesc. Psychiatry Ment. Health 2020, 14, 27. [CrossRef]
46. Rusticus, S.A.; Lovato, C.Y. Impact of sample size and variability on the power and type I error rates of equivalence tests:
A simulation study. Pract. Assess. Res. Eval. 2014, 19, 11. [CrossRef]
47. Jin, X.; Chen, W.; Sun, I.Y.; Liu, L. Physical health, school performance and delinquency: A comparative study of left-behind and
non-left-behind children in rural China. Child Abus. Negl. 2020, 109, 104707. [CrossRef] [PubMed]
48. Lan, X.; Wang, W. Is early left-behind experience harmful to prosocial behavior of emerging adult? The role of parental autonomy
support and mindfulness. Curr. Psychol. 2020, 1–14. [CrossRef]
49. Sun, X.; Tian, Y.; Zhang, Y.; Xie, X.; Heath, M.A.; Zhou, Z. Psychological development and educational problems of left-behind
children in rural China. Sch. Psychol. Int. 2015, 36, 227–252. [CrossRef]
50. Lan, X.; Marci, T.; Moscardino, U. Parental autonomy support, grit, and psychological adjustment in Chinese adolescents from
divorced families. J. Fam. Psychol. 2019, 33, 511–520. [CrossRef]
51. Cohen, J. Statistical power analysis. Curr. Dir. Psychol. Sci. 1992, 1, 98–101. [CrossRef]
52. Bradburn, N.M. Bradburn Scale of Psychologic Well-Being. In The Structure of Psychological Well-Being; Aldine: Chicago, IL, USA, 1969.
53. Chen, W.; Zhang, J. Factorial and construct validity of the Chinese positive and negative affect scale for student. Chin. Ment.
Health J. 2004, 18, 763–765.
54. Huebner, E.S. Preliminary development and validation of a multidimensional life satisfaction scale for children. Psychol. Assess.
1994, 6, 149–158. [CrossRef]
55. Tian, L.; Liu, W. Test of the Chinese version of multidimensional students’ life satisfaction scale. Chin. Ment. Health J. 2005,
19, 301–303.
56. Leung, A.N.M.; Wong, S.S.F.; Wong, I.W.Y.; McBride-Chang, C. Filial piety and psychosocial adjustment in Hong Kong Chinese
early adolescents. J. Early Adolesc. 2010, 30, 651–667. [CrossRef]
57. Yang, Y.; Li, P.; Kou, Y. Orientations to Happiness and Subjective Well-Being in Chinese Adolescents. Child Indic. Res. 2017,
10, 881–897. [CrossRef]
58. Busseri, M.A.; Sadava, S.W. A review of the tripartite structure of subjective well-being: Implications for conceptualization,
operationalization, analysis, and synthesis. Personal. Soc. Psychol. Rev. 2011, 15, 290–314. [CrossRef]
59. Jiang, J.; Song, Y.; Ke, Y.; Wang, R.; Liu, H. Is Disciplinary Culture a Moderator Between Materialism and Subjective Well-being?
A Three-Wave Longitudinal Study. J. Happiness Stud. 2016, 17, 1391–1408. [CrossRef]
60. Mantzicopoulos, P.; Neuharth-Pritchett, S. Development and validation of a measure to assess head start children’s appraisals of
teacher support. J. Sch. Psychol. 2003, 41, 431–451. [CrossRef]
61. John, O.P.; Donahue, E.M.; Kentle, R. The ‘Big-Five’ Inventory-Versions 4a and 54; Technical Report; Institute of Personality and
Social Psychology, University of California: Berkeley, CA, USA, 1991.
62. Carciofo, R.; Yang, J.; Song, N.; Du, F.; Zhang, K. Psychometric evaluation of Chinese-language 44-item and 10-item big five
personality inventories, including correlations with chronotype, mindfulness, and mind wandering. PLoS ONE 2016, 11, e0149963.
[CrossRef]
Int. J. Environ. Res. Public Health 2022, 19, 4103 16 of 16

63. Zhao, J.; Li, Q.; Wang, L.; Lin, L.; Zhang, W. Latent Profile Analysis of Left-behind Adolescents’ Psychosocial Adaptation in Rural
China. J. Youth Adolesc. 2019, 48, 1146–1160. [CrossRef]
64. Bradley, R.H.; Corwyn, R.F. Socioeconomic status and child development. Annu. Rev. Psychol. 2002, 53, 371–399. [CrossRef]
65. Batz, C.; Tay, L. Gender differences in subjective well-being. In Handbook of Well-Being; Diener, E., Oishi, S., Tay, L., Eds.; DEF
Publishers: Salt Lake City, UT, USA, 2018.
66. Schuessler, K.; Hittle, D.; Cardascia, J. Measuring Responding Desirably with Attitude-Opinion Items. Soc. Psychol. 1978,
41, 224–235. [CrossRef]
67. IBM Corp. IBM SPSS Statistics for Windows, Version 21.0; IBM Corp: Armonk, NY, USA, 2012.
68. Muthén, L.K.; Muthén, B.O. Mplus User’s Guide: Statistical Analysis with Latent Variables, 7th ed.; Muthén & Muthén: Los Angeles,
CA, USA, 2012.
69. The Jamovi Project. Jamovi. (Version 2.3) [Computer Software]. 2022. Available online: https://www.jamovi.org (accessed on
1 January 2022).
70. Little, R.J. A test of missing completely at random for multivariate data with missing values. J. Am. Stat. Assoc. 1988,
83, 1198–1202. [CrossRef]
71. Aiken, L.S.; West, S.G. Multiple Regression: Testing and Interpreting Interactions; Sage Publications: Thousand Oaks, CA, USA, 1991.
72. Maxwell, S.E.; Kelley, K.; Rausch, J.R. Sample size planning for statistical power and accuracy in parameter estimation. Annu. Rev.
Psychol. 2008, 59, 537–563. [CrossRef]
73. Marsh, H.W.; Lüdtke, O.; Trautwein, U.; Morin, A.J. Classical latent profile analysis of academic self-concept dimensions: Synergy
of person-and variable-centered approaches to theoretical models of self-concept. Struct. Equ. Model. A Multidiscip. J. 2009,
16, 191–225. [CrossRef]
74. Gu, X.; Yeung, W.J.J. Hopes and Hurdles: Rural Migrant Children’s Education in Urban China. Chin. Sociol. Rev. 2020, 52, 199–237.
[CrossRef]
75. Fu, T.M. Unequal primary education opportunities in rural and urban China. China Perspect. 2005, 2005. [CrossRef]
76. Suldo, S.M.; Friedrich, A.A.; White, T.; Farmer, J.; Minch, D.; Michalowski, J. Teacher support and adolescents’ subjective
well-being: A mixed-methods investigation. Sch. Psychol. Rev. 2009, 38, 67–85. [CrossRef]
77. Chen, Y.; Kubzansky, L.D.; VanderWeele, T.J. Parental warmth and flourishing in mid-life. Soc. Sci. Med. 2019, 220, 65–72.
[CrossRef]
78. McCrae, R.R.; Sutin, A.R. Openness to experience. In Handbook of Individual Differences in Social Behavior; Leary, M.R., Hoyle, R.H.,
Eds.; Guilford: New York, NY, USA, 2009; pp. 257–273.
79. Chen, X.; He, J.; Fan, X. Relationships between openness to experience, cognitive flexibility, self-esteem, and creativity among
bilingual college students in the US. Int. J. Biling. Educ. Biling. 2019, 215, 342–354. [CrossRef]
80. Shi, B.; Dai, D.Y.; Lu, Y. Openness to experience as a moderator of the relationship between intelligence and creative thinking:
A study of Chinese children in urban and rural areas. Front. Psychol. 2016, 7, 641. [CrossRef]
81. Migliorini, L.; Tassara, T.; Rania, N. A Study of Subjective Well-Being and Life Satisfaction in Italy: How are Children doing at
8 years of Age? Child Indic. Res. 2019, 12, 49–69. [CrossRef]
82. Jackson, J.J.; Hill, P.L.; Payne, B.R.; Roberts, B.W.; Stine-Morrow, E.A.L. Can an old dog learn (and want to experience) new tricks?
Cognitive training increases openness to experience in older adults. Psychol. Aging 2012, 27, 286–292. [CrossRef]
83. Stieger, M.; Wepfer, S.; Rüegger, D.; Kowatsch, T.; Roberts, B.W.; Allemand, M. Becoming more conscientious or more open to
experience? Effects of a two-week smartphone-based intervention for personality change. Eur. J. Personal. 2020, 34, 345–366.
[CrossRef]
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.